def main():
directory = sys.argv[1]
labels_file = sys.argv[2]
w = int(sys.argv[3])
h = int(sys.argv[4])
r = int(sys.argv[5])
print("Reading files")
#0. Get the names of all the files
list_of_files=getFileList(directory)
#0. Read the files into an OrderedDict datastructure.
datastructures=list()
for file_p in list_of_files:
datastructures.append(Collection(file_path=file_p, data_structure=readFile(file_p)))
print("Normalizing datastructures")
#1. (1a)For each data file, normalize data to values between 0.0 and 1.0
for collection in datastructures:
data_structure=collection.data_structure
normalize(data_structure)
#2. (1b)Determine the lengths and the ranges for the different bands.
bands=calculateBands(r)
#3. (1b) For each data file, change each of the values by the value of the center
#of the band to which that value corresponds.
for pair in datastructures:
datastructure=pair[1]
switch_values(datastructure, bands);
print("Creating training and testing sets")
#4. Read labels
file_label_l = readLabels(directory, labels_file)
(training_observations, testing_observations) = createObservations(datastructures, file_label_l)
print("training decision tree")
decision_tree_root = decisiontree.train(training_observations)
print("testing decision tree")
predictions = decisiontree.predict(decision_tree_root, testing_observations)
print(predictions)
#DO NOT USE THE WINDOW CREATION CODE
'''
def main():
directory = sys.argv[1]
labels_file = sys.argv[2]
w = int(sys.argv[3])
h = int(sys.argv[4])
r = int(sys.argv[5])
print("Reading files")
#0. Get the names of all the files
list_of_files = getFileList(directory)
#0. Read the files into an OrderedDict datastructure.
datastructures = list()
for file_p in list_of_files:
datastructures.append(
Collection(file_path=file_p, data_structure=readFile(file_p)))
print("Normalizing datastructures")
#1. (1a)For each data file, normalize data to values between 0.0 and 1.0
for collection in datastructures:
data_structure = collection.data_structure
normalize(data_structure)
#2. (1b)Determine the lengths and the ranges for the different bands.
bands = calculateBands(r)
#3. (1b) For each data file, change each of the values by the value of the center
#of the band to which that value corresponds.
for pair in datastructures:
datastructure = pair[1]
switch_values(datastructure, bands)
print("Creating training and testing sets")
#4. Read labels
file_label_l = readLabels(directory, labels_file)
(training_observations,
testing_observations) = createObservations(datastructures, file_label_l)
print("training decision tree")
decision_tree_root = decisiontree.train(training_observations)
print("testing decision tree")
predictions = decisiontree.predict(decision_tree_root,
testing_observations)
print(predictions)
#DO NOT USE THE WINDOW CREATION CODE
'''
def predict():
try:
tree_json = request.json["tree"]
toy_id = request.json["toy_id"]
user_truth = request.json["user_truth"]
targets = request.json["category"]
tree = utilities.get_tree(tree_json)
y_pred = decisiontree.predict(tree, toy_id, user_truth, targets[0])
print('Prediction: ', targets[y_pred[0]])
return {"prediction": y_pred.tolist()}
except Exception as err:
print(err)
return "Failed", 500
def task_list(request):
Method = request.method
if request.method == 'GET':
#serializer = TaskSerializer(data=request.data)
return render(request, 'snippets/project.html', locals())
elif request.method == 'POST':
postBody = request.body
json_result = json.loads(postBody)
if json_result['type']=='delConnection':
sourceid=json_result['sourceid']
targetid=json_result['targetid']
finalConnections(line, sourceid, targetid)
resp={
'code' : '1'
}
return HttpResponse(json.dumps(resp),content_type="application/json")
if json_result['type']=='saveConnection':
relId = json_result['relId']
taskid = json_result['task.taskId']
totaskId = json_result['toTaskId']
sType = json_result['sourceType']
tType = json_result['targetType']
listtype = []
listtype.append(sType)
listtype.append(tType)
listid = []
listid.append(taskid)
listid.append(totaskId)
connections = {}
connections['cid'] = relId
connections['connids'] = listid
connections['conntype'] = listtype
line.append(connections)
resp={}
return HttpResponse(json.dumps(resp),content_type="application/json")
if json_result['type']=='delNode':
endid=json_result['endid']
finalNode(line, endid)
resp={}
return HttpResponse(json.dumps(resp),content_type="application/json")
if json_result['type']=='saveLine':
endid=json_result['endid']
endtype=json_result['endtype']
finalline(line,endid,endtype)
resp={}
return HttpResponse(json.dumps(resp),content_type="application/json")
if json_result['type']=='data_source':
global filename
filename = json_result['data']
with open('E:/Anaconda/Scripts/CorsApi/snippets/Resource/file/'+json_result['data']+'.csv','rt',encoding="utf-8") as csvfile:
reader = csv.reader(csvfile)
rows = [row for row in reader]
csvfile.close()
global row
row = rows[0]
resp={
'attribute':row,
}
return HttpResponse(json.dumps(resp),content_type="application/json")
if json_result['type']=='field':
column_data = {}
for attribute in json_result['data']:
with open('E:/Anaconda/Scripts/CorsApi/snippets/Resource/file/'+filename+'.csv','rt',encoding="utf-8") as csvfile:
inner_list=[]
reader = csv.DictReader(csvfile)
column = [row[attribute] for row in reader]
if attribute!='result':
X = np.array(column,dtype=np.float64).reshape(-1,1)
min_max_scaler = preprocessing.MinMaxScaler()
X_minMax = min_max_scaler.fit_transform(X)
for n in X_minMax:
inner_list.append(round(n[0],2))
column_data[attribute]=inner_list
else:
rs=map(eval,column)
column_data[attribute]=list(rs)
resp={
'normalization':column_data,
}
return HttpResponse(json.dumps(resp),content_type="application/json")
if json_result['type']=='normal':
column_data = {}
data = json_result['data']
for attribute in data:
if attribute!='result':
X = np.array(column,dtype=np.float64).reshape(-1,1)
min_max_scaler = preprocessing.MinMaxScaler()
X_minMax = min_max_scaler.fit_transform(X)
for n in X_minMax:
inner_list.append(round(n[0],2))
column_data[attribute]=inner_list
else:
rs=map(eval,column)
column_data[attribute]=list(rs)
resp={
'normalization':column_data,
}
return HttpResponse(json.dumps(resp),content_type="application/json")
if json_result['type']=='pca':
data = json_result['data']
X = np,array(data)
pca = PCA(n_components=2)
ratio = pca.explained_variance_ratio_
resp={
'ratio':ratio[0],
}
return HttpResponse(json.dumps(resp),content_type="application/json")
if json_result['type']=='bayes':
data = json_result['data']
dataset,trainingSet,testSet,accuracy,result = naivebayes.bayes(data)
attribute=[]
for key in data.keys():
attribute.append(key)
resp={
'rows':len(dataset),
'train':len(trainingSet),
'test':len(testSet),
'accuracy':accuracy,
'result':result,
'attribute':attribute,
}
return HttpResponse(json.dumps(resp),content_type="application/json")
if json_result['type']=='decision_tree':
max_depth = int(json_result['max_depth'])
criterion = json_result['criterion']
splitter = json_result['splitter']
min_samples_split = int(json_result['min_samples_split'])
data = json_result['data']
score = decisiontree.dtree(max_depth,criterion,splitter,min_samples_split,data)
resp={
'score':score,
}
return HttpResponse(json.dumps(resp),content_type="application/json")
if json_result['type']=='dtree_predict':
a=[]
dtree_predict= json_result['data']
#dtree_predict = np.float64(dtree_predict)
a.append(dtree_predict)
pred = decisiontree.predict(a)
resp={
'pred':[str(i) for i in pred],
}
return HttpResponse(json.dumps(resp),content_type="application/json")
return render(request, 'snippets/project.html', locals())
print(output)
assert output != 0
####################################################################################################
# Test predict()
####################################################################################################
print("\n\nTesting predict()")
print("----------------------------------------------------------------------------------------------------\n\n")
idSubtrees = {0: 0, 1: 1}
notSubtrees = {0: 1, 1: 0}
data = [[1], [1], [0], [0]]
tree = 3
output = dt.predict(tree, data)
print(output)
assert output == [3, 3, 3, 3]
data = [[1], [1], [0], [0]]
tree = (0, {0: 10, 1: 20})
output = dt.predict(tree, data)
print(output)
assert output == [20, 20, 10, 10]
data = [[1, 0], [1, 0], [0, 0], [0, 0]]
tree = (0, {0: 10, 1: 20})
output = dt.predict(tree, data)
print(output)
assert output == [20, 20, 10, 10]
def random_predict(treesset, testdata, features, label):
predicts = []
for tree in treesset:
predict = dt.predict(tree, features, testdata, label)
predicts.append(predict)
return majorvoting(predicts)
def predict(forest: RandomForest, row: Row) -> Label:
predictions = [decisiontree.predict(tree.tree, row) for tree in forest]
return util.most_frequent_alphabetically(predictions)